Message-ID: <cd445f29-ea01-4e57-b1b7-4e9c3bfbcc8f@kernel.org>
Date: Wed, 19 Mar 2025 08:39:52 +0100
From: Krzysztof Kozlowski <krzk@...nel.org>
To: Christian Marangi <ansuelsmth@...il.com>
Cc: Michael Turquette <mturquette@...libre.com>,
 Stephen Boyd <sboyd@...nel.org>, Rob Herring <robh@...nel.org>,
 Krzysztof Kozlowski <krzk+dt@...nel.org>, Conor Dooley
 <conor+dt@...nel.org>, Lee Jones <lee@...nel.org>,
 Vinod Koul <vkoul@...nel.org>, Kishon Vijay Abraham I <kishon@...nel.org>,
 Matthias Brugger <matthias.bgg@...il.com>,
 AngeloGioacchino Del Regno <angelogioacchino.delregno@...labora.com>,
 Greg Kroah-Hartman <gregkh@...uxfoundation.org>,
 Lorenzo Bianconi <lorenzo@...nel.org>, Daniel Danzberger <dd@...edd.com>,
 Arnd Bergmann <arnd@...db.de>, Linus Walleij <linus.walleij@...aro.org>,
 Nikita Shubin <nikita.shubin@...uefel.me>, Guo Ren <guoren@...nel.org>,
 Yangyu Chen <cyy@...self.name>, Ben Hutchings <ben@...adent.org.uk>,
 Felix Fietkau <nbd@....name>, linux-clk@...r.kernel.org,
 devicetree@...r.kernel.org, linux-kernel@...r.kernel.org,
 linux-arm-kernel@...ts.infradead.org, linux-phy@...ts.infradead.org,
 linux-mediatek@...ts.infradead.org, linux-usb@...r.kernel.org,
 upstream@...oha.com
Subject: Re: [PATCH 09/13] dt-bindings: phy: Add documentation for Airoha
 AN7581 USB PHY

On 18/03/2025 12:31, Christian Marangi wrote:
> On Sun, Mar 16, 2025 at 06:01:02PM +0100, Krzysztof Kozlowski wrote:
>> On 11/03/2025 19:51, Christian Marangi wrote:
>>>>
>>>>> +    $ref: /schemas/types.yaml#/definitions/uint32
>>>>> +    enum: [0, 1]
>>>>
>>>> I don't understand why do you need index property here (which are
>>>> usually not allowed).
>>>>
>>>
>>> Eh... As said in the description this is really to differentiate the 2
>>> different physical port...
>>>
>>> Each port have a dedicated oscillator for calibration and these
>>> calibration are identified by an offset (all placed one after another in
>>> a separate register space).
>>
>> So different oscillators? Then describe the oscillator and its differences.
>>
>> Different programing model? Different compatible.
>>
>> Other difference? Depending what is the difference.
>>
>> But there is no such thing as "different port ID" based on your
>> description above. You just claimed that they are different, but you do
>> not put that difference to hardware description. Instead you encode that
>> difference in the drivers and it should be opposite. The DTS, so the
>> hardware description, should tell you the difference. And I am sorry,
>> but in 99% of cases "I am the first phy" and "I am the second" is not
>> the actual difference we are interested in.
>>
> 
> Ok to make it as clear as possible. (hope I don't contraddict with the
> previous statement) (actually yes the "separate register space statement
> was wrong and sorry for the confusion")
> 
> - 2 USB port
> - USB 2.0 needs to be calibrated with an oscillator
> - Each USB port have his own dedicated oscillator somewhere in the HW
> - Each USB port have at the same offset a register to SELECT the
>   oscillator. This register refer to the same oscillator selection in
>   the HW.
>   
>   Example:
>   reg 0x1fac0100 MASK 27:26 can be set to source out of oscillator 0,1,2
>   reg 0x1fae0100 MASK 27:26 can be set to source out of oscillator 0,1,2

So that's your difference between devices - you choose the oscillator.

> 
>   Both register refer to the same oscillators in hardware (hence each
>   port should select the correct one)
>   Selecting oscillator 0 for both USB port is problematic.
> 
> With this in mind is it ok if I describe this with something like
> 
> airoha,usb2-monitor-clk-sel = <AIROHA_USB2_MONCLK_SEL0>;
> 
> and some dt-bindings include
> 
> #define AIROHA_USB2_MONCLK_SEL0 0
> #define AIROHA_USB2_MONCLK_SEL1 1
> #define AIROHA_USB2_MONCLK_SEL2 2
> #define AIROHA_USB2_MONCLK_SEL3 3
> 
> Or an enum of string like "osc0", "osc1", "osc2"...?

Both are fine.

Best regards,
Krzysztof

Powered by blists - more mailing lists